A cloud-based, automated surface deformation mapping framework using artificial neural networks and radar interferometry: Steps toward commercializing RADARSAT Constellation Mission data - NS-017

Genre de projet: Recherche
Discipline(s) souhaitée: Génie - autre, Génie, Informatique, Sciences mathématiques, Sciences de l’environnement, Sciences naturelles, Géographie / géologie / sciences de la terre, Océanographie
Entreprise: KorrAI Technologies Ltd.
Durée du projet: Flexible
Preferred start date: As soon as possible.
Langue exigée: Flexible
Emplacement(s): Halifax, NS, Canada; Andorre; Canada
Nombre de postes: 1
Niveau de scolarité désiré: MaîtriseRecherche postdoctorale
Rechercher dans les réseaux internationaux de Mitacs - cochez cette case si vous souhaitez recevoir des profils de chercheurs basés à l’extérieur du Canada: 
Yes

Au sujet de l’entreprise: 

KorrAI is a Halifax-based technology company that uses satellite imagery & AI enables mining companies to accelerate exploration timelines while maintaining a solid brand of environmental sustainability. Focussing on critical minerals, they are decrypting the tangled web of EO data and helping secure the multi-billion critical mineral supply chain.

Veuillez décrire le projet.: 

Radar satellites can penetrate clouds and provide reliable imagery in all weather conditions. A processing method, known as radar interferometry, or InSAR, can produce surface deformation maps and detect deformation change through time. KorrAI is combining this method with cloud computing and artificial intelligence to provide an efficient framework to automatically detect changes in surface movement.

Classic InSAR products, including competing solutions, are offered as static map products. With the Government of Canada’s RADARSAT Constellation Mission (RCM) repeat interval of 4 days in Canada, KorrAI’s solution will provide near real-time, dynamic surface deformation monitoring.

KorrAI in partnership with the Canadian Space Agency is to commercialize the petabytes of over twenty years’ worth of historical radar data collected from the previous RADARSAT-1 mission, as well as ongoing data collected by the RCM mission, providing Canadian firms with an unprecedented level of data access and analytical capabilities.

Main tasks to be performed by the candidate:

  • Write clean, high-quality, maintainable code
  • Develop new approaches to improve current production processes and end-client products.
  • Develop and oversee the application of deep learning models to InSAR data pipelines
  • Keep engaged in the latest research for new areas of algorithm and product development
  • Manage junior members in the remote sensing team, collaborate with other team members and stakeholders
  • Coordinate cross-functionally to ensure the project meets business objectives and compliance standards.
  • Support writing proposals and generating winning bids for future InSAR

Methodology/techniques that will be used:

  • Sentinel, RCM, Radarsat data
  • GCP
  • Google Compute Cloud (GCP)
  • ESA’s Sentinel Application Platform (SNAP) or GMTSA
  • InSAR Scientific Computing Environment (ISCE) 

 

Expertise ou compétences exigées: 

  • Masters or Postdoctoral researcher in GIS, earth sciences, or other related degrees
  • Knowledge and experience working with Satellite Synthetic Aperture Radar (SAR) (RADARSAT-1 and -2; Sentinel-1, airborne systems) is an essential
  • Knowledge of Interferometric analysis and its use within an applied application area
  • Thorough, demonstrated knowledge of open-source GIS tools such as QGIS and GDAL